Spatiotemporal Distribution of Zika Virus and Its Spatially Heterogeneous Relationship with the Environment

Round 1

Reviewer 1 Report

The authors have modeled the recent Zika virus outbreak in S. American and worldwide looking at factors such as GDP, environment, population density, landscape, etc. This is a broad scope and as such lacks the detail in any area to be convincing that these data can be used for analysis of future outbreaks of ZIKV or SARS-CoV-2.

The authors have COVID-19 discussed throughout the paper; they should remove all references to COVID-19 and focus on ZIKV or change the overall focus of their paper.

There are many broad statements made in this paper that do not seem to have data to support it. Better to have a well supported claim than to make a number of not so well supported claims - in my opinion.

The authors need to more clearly discuss ZIKV - is it a "wild animal"?; they say they did not consider the A. aegypti density in this study - that is a major factor in spread; they need to discuss what they feel is a "highly contagious arbovirus" and how population density is the key factor to spread; they specifically mention closer person-to-person contact (maternal, sexual transmission?).   

Author Response

We are pleased to resubmit the revised version of manuscript ijerph-1014649, entitled “Spatiotemporal distribution of Zika virus and its spatially heterogeneous relationship with the environment”. We gratefully thanks for the valuable comments that have greatly helped us to improve our paper. The specific comments are laid out below. Our responses to each comment and changes/additions to the manuscript are given in blue text. We hope that the revised version is now suitable for publication.

Reviewers' comments:

Response: We gratefully thank the reviewer for his/her outstanding comments, which greatly helped us to improve the quality of our manuscript. Below, we address every comment carefully and explain the corresponding changes in the manuscript.

1. The authors have modeled the recent Zika virus outbreak in S. American and worldwide looking at factors such as GDP, environment, population density, landscape, etc. This is a broad scope and as such lacks the detail in any area to be convincing that these data can be used for analysis of future outbreaks of ZIKV or SARS-CoV-2.

Response: Thank you for the helpful comment. The availability of detail data or the substitutability of detailed information should also be considered when selecting factors. Therefore, after considering these factors and referring to relevant literature, the environmental factors, including climate, ecology and society, were determined, according to the possible influence of environmental factors on the spread of ZIKV. We have revised the manuscript in lines 118 to 129 as “Weather and climate change influence mosquito geographic distribution, population abundance, lifespan, and transmission potential [24]. The research shows that the temperature range of ZIKV transmission is 18°~34°, and it peaks at 29° [25]. In the context of global warming, the annual temperature is likely to rise to the optimal value for ZIKV transmission. In 2015, Brazil’s winter and spring were the warmest and driest on record, with higher winter temperatures contributing to ZIKV's propagation and spread [26]. Studies have shown that ZIKV-linked microcephaly cases has an obvious correlation with the percentage of forest coverage [26]. The high density of forest cover was negatively correlated with the prevalence rate of ZIKV, which could be interpreted as deforestation and destruction of habitat led to a deterioration of landscape fragmentation. Meanwhile, population density changes from a lower level to a higher level, expanding the risk of ZIKV transmission. In addition, in terms of socio-ecological factors, studies have pointed out that higher GDP and health conditions reduce the transmission of ZIKV [27].”

In addition, this study can analyze the transmission risks of ZIKV in different regions from the perspective of geography. Therefore, based on the dominant environmental factors, corresponding measures can be proposed to prevent the further transmission of ZIKV. We have revised manuscript in lines 301 to 316 as “As clinical trials of ZIKV vaccines are still underway, more efforts are needed to develop antivirals for ZIKV, and reduce the likelihood of another ZIKV outbreak [37]. Based on the concept of “classification-coordination-collaboration” [38,39], the systematic approach towards fighting ZIKV can be proposed. Authorities should increase vigilance and surveillance toward imported cases of ZIKV infection which would most possibly reduce autochthonous ZIKV transmission and global spread according to the transmission route in 2016 (Figure 7). Moreover, the personal education should be greatly emphasized to make them aware of the potential ways of ZIKV transmission and preventative measures. Based on the dominant factors impacting different regions (Figure 8d), corresponding measures should be taken based on environment-link level. For people in population density-caused areas, they should take precautions against mosquitoes, such as wearing light-colored long-sleeved shirts and trousers, and using insect repellent to exposed parts of their body to minimize vector contact [40]. In areas with ecological environment destructed most, people should pay attention to ecological protection, and relevant departments should carry out ecological management projects to restore the ecology. Besides, public health authorities in ZIKV endemic regions should provide standard healthcare precautions to solve the ZIKV transmission caused by the difference in GDP per capita level.”.

2. The authors have COVID-19 discussed throughout the paper; they should remove all references to COVID-19 and focus on ZIKV or change the overall focus of their paper.

Response: Sorry for unclear expression. We believed that this study can provide a methodological reference for analysis of the relationship between COVID-19 transmission and socio-ecological environmental factors. In other words, the relationship can be discussed from the perspective of geography. In addition, the study found that compared with the natural environment change, the change of social environment caused by human activities for the spread of ZIKV influence is bigger. The conclusions of this study may provide insights into the relationship between COVID-19 transmission and environmental factors. For zoonotic and highly infectious diseases such as ZIKV and COVID-19, changes in the natural environment may not be the key factor, but may be caused by changes in the social environment for human activities. Therefore, we have deleted most of the COVID-19 content from the manuscript, and only maintained a few parts. The manuscript in lines 85 to 88 has been revised as “As the SARS-CoV-2 pandemic continues to spread, this study can provide methodological experiences for exploring the relationship between SARS-CoV-2 and environmental factors to accelerate the fight against SARS-CoV-2 and achieve the goal of sustainable development sooner” and the lines 317 to 323 as “The analysis of ZIKV shows, patterns of SARS-CoV-2 propagation can be developed and analyzed from a geographic perspective, considering that the heterogeneity of geographical and environmental conditions may cause differences in the transmission of SARS-CoV-2. In addition, by selecting a comprehensive set of environmental factors that influence the spread of SARS-CoV-2, such as the landscape fragmentation, biodiversity index and GDP index selected in this study, the source of the virus can be determined at various levels and the mechanisms driving the influence of environmental factors on the spread of SARS-CoV-2 can be identified.” and the lines 343 to 347 as “Therefore, the conclusions of this study can also provide methodological references for studying the relationship between the SARS-CoV-2 pandemic and the environment. In other words, compared with changes in the natural environment, the impact of human activities plays a vital role in the emergence and development of infectious diseases.”

3. There are many broad statements made in this paper that do not seem to have data to support it. Better to have a well-supported claim than to make a number of not so well supported claims - in my opinion.

Response: Thank you for your constructive comment. In view of the incomplete analysis of ZIKV spatiotemporal distribution in this study and the lack of corresponding data analysis in some conclusions, we adopted the method of combining exploratory spatial data analysis (ESDA) with seasonal index, and further modified the statement and data analysis methodologically. We have revised the manuscript in lines 169 to 190 and added a section (2.3.1) named “Spatiotemporal analysis of ZIKV epidemic”, described as “Exploratory spatial data analysis (ESDA) is used to analyze the spatiotemporal distribution of ZIKV epidemic. Based on GIS technology, spatial analysis of geographic information is combined with the spread of infectious diseases to explore the spatial distribution pattern of ZIKV. In this study, the ESDA-GIS analysis model was used to explore the spatial clustering and differentiation of ZIKV transmission from the perspectives of hotspot analysis and spatial correlation analysis, and reveal the rule of its spatiotemporal distribution. Hotspot analysis (Getis-Ord Gi *) can express the features of cold spots and hot spots in a certain regional spatial clustering, and intuitively display the hot spots and cold spots within the scope of the research area. It identifies statistically the clusters of points having higher magnitude using the quantity of Relative Risk Ratio [31]. Moran’s Index (Moran’s I) statistics was used to evaluate autocorrelation in ZIKV spatial distribution and test how regions were clustered or dispersed in space [32]. The LISA cluster index is the indicator of spatial association. The localities that show statistically significant (0.05 level) cluster of high values are indicated by (HH) and the localities that show statistically significant (0.05 level) cluster of low values are indicated by (LL). A low negative z score (for example, z <−1.96) for a locality indicates if the locality has a high value and is surrounded by locality with low values (HL) or if the locality has a low value and is surrounded by localities with high values (LH) [31].Seasonal index is a method to describe the seasonal change of time series based on the characteristic of seasonal cycle variation. The seasonal index method analyzes and forecasts the data according to the seasonal regularity. The SPSS v26.0 (IBM SPSS Statistics v26) software is used to calculate the seasonal index of ZIKV from 2015 to 2018.”

We also have revised the manuscript in lines 252 to 255 as “Pearson correlation coefficient was used to analyze the correlation between environmental factors and confirmed ZIKV cases (Table 1). The result indicated that GDP per capita, biodiversity and relative humidity had a negative affect whereas temperature, population density and landscape fragmentation had a positive effect.”. Table 1, the manuscript in line 277 is as follows:

Table 1. The correlation coefficient of different environmental factors.

Indicators	GDP per capita	Population density	Temperature	Relative humidity	Biodiversity	Landscape fragmentation
Pearson correlation coefficient	-0.047	0.266	0.035	-0.058	-0.133	0.105

 

4. The authors need to more clearly discuss ZIKV - is it a "wild animal"?; they say they did not consider the A. aegypti density in this study - that is a major factor in spread; they need to discuss what they feel is a "highly contagious arbovirus" and how population density is the key factor to spread; they specifically mention closer person-to-person contact (maternal, sexual transmission?).

Response: Thank you for the helpful comment. According to the discussion of ZIKV spread, we have revised the manuscript in lines 38 to 39 as “ZIKV is an emerging arthropod-borne virus (arbovirus) belonging to the family Flaviviridae and genus Flavivirus [5].” and lines 47 to 50 as “ZIKV is transmitted primarily by the bite of infected mosquitoes which are mainly Aedes aegypti and Aedes albopictus [7]. A. aegypti mosquitoes are widely distributed in the tropical and sub-tropical regions, whereas A. albopictus mosquitoes are widely distributed throughout the tropical, sub-tropical and temperate regions [8].” We attempted to obtain the global distribution data of A. aegypti and A. albopictus, and found that the distribution of mosquito vectors was predicted by the species distribution model (SDM) using the environmental variables. However, the environmental variables in the SDM model were also the environmental factors in this study, and there would be an overlap in data utilization if we used the mosquito distribution data. In addition, the environmental factors selected in this study are based on the fact that they have an indirect influence on ZIKV spread, directly affecting the distribution of mosquito vectors. Therefore, the distribution of A. aegypti and A. albopictus was not taken into account in this model.

For the discussion of "highly contagious arbovirus", we have revised the manuscript in lines 233 to 241 as “The analysis of the spatial distribution of ZIKV outbreaks over time (Figure 7) shows that the ZIKV outbreak mainly occurred in the tropics around the equator and then the epidemic rapidly spread to many countries and regions around the world, including North America, Europe, Asia and Southeast Asia in 2016，which indicates that the transmission of ZIKV originates from and not limits to A. aegypti and A. albopictus. With the increasing progress of globalization and global warming, ZIKV is spreading more and more widely around the world. As a highly contagious arbovirus, ZIKV is highly contagious, which can be further spread, particularly via viremic travelers or the movement of infected mosquitoes [34]. Moreover, autochthonous ZIKV transmission as well as cases of ZIKV transmission via physical activity were also been reported [35].”

We are sorry for the confusing description about population density. We have modified the explanation of population density in lines 278 to 281 as “The result shows that population density has the widest impact and is the key factor in the spread of ZIKV in South America. The reason may be that high population density increases the risk of people being bitten by infected mosquitoes, and thus spreading the virus worldwide. Domestic and international travel may aggravate the spread of the virus in this region (Figure 8d).”

Last but not least, we would like to express our sincere thanks again for all the outstanding comments and suggestions, which greatly helped us to improve the technical quality and presentation of our work.

 

Author Response File: Author Response.docx

Reviewer 2 Report

In this manuscript, Jie Li et al.demonstrated that social factors have a greater influence than natural factors on the spread of ZIKV by analyzing the environmental parameter of population density, GDP per capita, and landscape fragmentation using the method of GWR analysis mode. I think the data analysis in the article is too simple, the social factors, such as religion, lifestyle, or living habits should also be included in the model for more convictive analysis. Furthermore, the type size in Figure 1,2,5,6,7 are too small to be clear and should be revised for better visual effect.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

In this revised manuscript, Jie Li et al. added some contents in the introduction, results, and discussion that make the article more logical and believable. For example, the temperature range of ZIKV transmission is mentioned and explained how global warming affects the climate and transmission of ZIKV. But some small problem should be revised:

1. In line 58, the verb of "led" should be "lead".
2. In Figure6, (a) is not clear and a little bit small and the font is different from (B) and (C).
3. In Figure8, the font of “Ocean” is different from the other three dominant factors.

Author Response

Please see the attachment.

Author Response File: Author Response.docx